metadata
dataset_info:
features:
- name: image
dtype: binary
- name: table
dtype: string
- name: sample_id
dtype: int64
splits:
- name: train
num_bytes: 10554587948
num_examples: 304997
download_size: 9889037806
dataset_size: 10554587948
configs:
- config_name: default
data_files:
- split: train
path: data/train-*